topic Re: Missing value of False Negative in Classification Table - Gradient Boosting Model - SAS EM in SAS Data Science

Missing value of False Negative in Classification Table - Gradient Boosting Model - SAS EM

dishen_pancholi — Sun, 17 Apr 2022 21:45:57 GMT

Hello Everyone,

I am using SAS Enterprise Miner to create a model. While using the Gradient Boosting model, I am getting missing value for false negative in the training set. I'm not sure what it means.

Moreover, in what ways can we improve the classifiers for the model?

Thanks,

Dishen

Re: Missing value of False Negative in Classification Table - Gradient Boosting Model - SAS EM

sbxkoenk — Mon, 18 Apr 2022 14:43:21 GMT

I have moved this post to 'SAS Data Science' board which is more appropriate for the subject.

Re: Missing value of False Negative in Classification Table - Gradient Boosting Model - SAS EM

sbxkoenk — Mon, 18 Apr 2022 14:49:47 GMT

Hello,

How many observations are there in your training dataset?

If you have 234 + 3 + 80 ( = TN + FP + TP ) = 317, then the missing means zero!

Making the model better?

The biggest gain is in the data you use (include more relevant candidate predictors).
The 2nd biggest gain is in how you deal with this data (dimension reduction, feature construction, variable transformations, …).
Only little gain (typically) is in considering another model (different from your Gradient Boosting model).

Remember : what you do BEFORE throwing the data into a modelling node is of utmost importance !

Good luck,

Koen

Re: Missing value of False Negative in Classification Table - Gradient Boosting Model - SAS EM

WendyCzika — Mon, 18 Apr 2022 18:58:25 GMT

You might find some helpful tips for improving the gradient boosting model here:

https://communities.sas.com/t5/SAS-Communities-Library/Quick-tips-for-setting-your-Gradient-Boosting-node-properties-in/ta-p/513600